📊 MGF Dataset Explorer (Shiny App)

An interactive Shiny for Python web app for browsing and analyzing climate-related policy actions by central banks and financial regulators — developed by the Macro and Green Finance Lab at Peking University (MGF@PKU).

This tool is designed as a template for research, presentation, and educational purposes. It is not intended for local hosting by the general public — instead, it showcases how one might integrate real-time GitHub-hosted datasets into a Shiny app.

---

🔍 Overview

This app visualizes a curated dataset of climate and sustainability-related regulatory measures from major economies (G20, EU, etc.) since 2021. Features include:

• 🔎 Filtering by region, policy type, year, and keyword
• 🧾 Structured detail view for each policy
• 🔗 Links to original sources
• 📥 Export of filtered results (CSV)

Powered by:

• shiny for interactivity
• polars for fast data wrangling
• GitHub API for real-time data retrieval
• Custom CSS + JS enhancements

---

📦 Features

• Paginated and stylized data table
• Clickable rows with detailed policy views
• CSV export of current filtered results
• Integrated GitHub data sync (via authenticated API)
• Modern UI with iconography, tooltips, and layout styling

---

🧪 Setup (For Contributors Only)

This app uses GitHub as a remote data backend. If you're customizing or extending the app:

1. Clone the repo

git clone https://github.com/MGFPKU/MGF_dataset_shiny.git
cd MGF_dataset_shiny

2. Install dependencies

We use uv for fast dependency resolution:

uv sync

3. Set up environment variables

⚠️ We do not publish a GitHub token in this repo. If you need access for development or testing, contact the maintainers for a personal access token (PAT) with read-only access to the dataset repository.

You’ll need to set the token in a .env file like:

GITHUB_TOKEN=ghp_...
GOOGLE_SCRIPT_URL=https://script.google.com/macros/s/...
LANGUAGE=EN

Below is what each variable does:

• GITHUB_TOKEN: a GitHub Personal Access Token (PAT) with read access to the dataset repository. The app uses this token when calling the GitHub API to fetch data/data.csv. Keep this token private (do not commit it).
• GOOGLE_SCRIPT_URL: the public URL for a Google Apps Script web app that acts as the mailing bot. The Shiny app POSTs filtered exports (CSV/XLSX) to this endpoint and the script forwards them by email.
• LANGUAGE: set the UI language for the app. Use EN for English or CN for Chinese. The value controls which translations are displayed in the interface.

4. Run the app

python -m shiny run --reload  app.py

Then open your browser to:

http://localhost:8000

Or more conveniently, install the shiny extension to Positron/VS Code and press the Run button.

---

📁 Project structure

The repository layout and purpose of key files:

.env                  # Environment variables (not committed, create locally following setup instructions)
app.py                # Main Shiny app (UI + server)
details.py            # Renders detailed policy view
table.py              # Paginated table output and helpers
download.py           # Download UI and mailing helpers (POSTs to Google Script)
i18n.py               # Translation helper; reads LANGUAGE to switch UI
translation.json      # Translation strings used by `i18n.py`
pyproject.toml        # Project metadata / build config (managed by uv)
.python-version       # Python version pinning for uv
requirements.txt      # Dependency pins (optional with pyproject)
LICENSE               # License
README.md             # Project documentation (this file)
.gitignore            # Git ignore rules

Notes:

• Edit app.py to change high-level UI or filtering logic.
• table.py and details.py control how policy rows and detail pages are rendered.
• download.py integrates with the Google Apps Script mailing bot (set via GOOGLE_SCRIPT_URL).

📚 Citation

If you use the MGF dataset in your research or writing, please cite us as follows:

Macro and Green Finance Lab, Peking University (MGF@PKU)
"Climate-related Policy Tracker for Central Banks and Regulators"
https://mgflab.nsd.pku.edu.cn/MGFsjk/zczz/index.htm, 2025.

---

📁 Data Source

The dataset is hosted in a private GitHub repository along with our scraping scripts. We apologize for the inconvenience. But the dataset provided through the app is updated in real-time and contains most information available on the original source.

---

🏛️ Credits

Developed by Dianyi Yang (@kv9898) for the Macro and Green Finance Lab at Peking University (MGF@PKU).

If you use or adapt this project for academic or institutional purposes, please cite the dataset accordingly and feel free to reach out for collaboration or inquiries.